AI agent went rogue and hacked startup by itself, OpenAI reveals - The Guardian

1. AI agent went rogue and hacked startup by itself, OpenAI reveals  The Guardian
2. OpenAI and Hugging Face partner to address security incident during model evaluation  OpenAI
3. OpenAI cyber models broke out of training environment to hack Hugging Face  CNBC
4. OpenAI says its AI went rogue and launched 'unprecedented' cyber-attack  BBC
5. An OpenAI test model escaped and broke into a real company’s servers  CNN